DTA

Digital Theses Archive

 

Tesi etd-10272022-123316

Type of thesis
Corso Ordinario Secondo Livello
Author
TONI, FRANCESCO
URN
etd-10272022-123316
Title
Pseudo Rational Agent-Based-Model Expectations
Structure
Cl. Sc. Sociali - Scienze Economiche
Course
SCIENZE ECONOMICHE E MANAGERIALI - SCIENZE ECONOMICHE E MANAGERIALI
Committee
relatore Prof. ROVENTINI, ANDREA
Relatore Prof.ssa Treibich Tania
Relatore Dott. VANDIN, ANDREA
Tutor Prof. BOTTAZZI, GIULIO
Presidente Prof. PICCALUGA, ANDREA MARIO CUORE
Membro Dott.ssa CANTARELLI, PAOLA
Membro Prof. NUVOLARI, ALESSANDRO
Membro Prof. MONETA, ALESSIO
Membro Prof. FAGIOLO, GIORGIO
Membro Prof. MINA, ANDREA
Keywords
  • agent-based computational economics
  • complexity
  • heterogeneity
  • rational expectations
  • recursive learning
Exam session start date
02/12/2022;
Availability
completa
Abstract
We enrich the modelling of expectations in agent-based-models by introducing Pseudo Rational ABM Expectations. Analogously to Muth&#39;s hypothesis, agents perform forecasting by using the &#34;true model&#34; of the economy. More precisely, the novel expectations are given by a recursive forecasting method, whereby the agents, aware of the feedback loop between expectations and the system dynamics, revise their expectations iteratevely by foreseeing the future state variables at each step.<br><br>The new expectations are implemented in the Keynes meeting Schumpeter model and compared with static naïve expectations. We find that the sophisticated expectations imply perfect foresight, boosting the firm-level performance of the adopters. Nevertheless, introducing one forward-looking firm only deteriorates the macroeconomic dynamics, which gets more volatile and unstable.
Files